CHEESY METAMORPHOSIS (TUBERCLE, CAXCER). 525 



occur which look exactly like tubercles. I have demon- 

 strated that it is by the gradual transformation of the 

 elements of cancer that this cheesy matter is produced. 

 But if we did not positively know from the history of 

 their development that cancer-cells disintegrate step by 

 step, and that no tubercles form in the middle of cancer, 

 we should in many cases be altogether unable to arrive 

 at any decision from merely examining the specimen. 



If those difficulties be surmounted which lie in the 

 external appearance of the formation, and lead the ob- 

 server astray not only when he considers its grosser 

 features, but also when he investigates its more intimate 

 composition, there remains nothing else to assist us in 

 coming to a right conclusion than the investigation of the 

 type of development displayed by the individual new-for- 

 mations during the stages of their actual development, 

 not during those of their retrograde metamorphosis. 

 The nature of tubercle cannot be studied after the period 

 when it becomes cheesy, for from that time its history is 

 identical with the history of pus which is becoming 

 cheesy ; an earlier period must be chosen when it is 

 really engaged in proliferation. So in the case of other 

 formations, that period must be studied which is com- 

 prised between their origin and their culminating point, 

 and we must see with what normal physiological types 

 they agree. Then it is, I think, certainly possible for us 

 to arrive at a just conclusion with the aid of the simple 

 principles of histological classification, which I have 

 already propounded to you (p. 91). Heterologous tissues 

 also have physiological types. 



A colloid growth, if we really take it to mean what 

 Laennec did a gelatinous organized new-formation 

 must necessarily correspond to some type to be met with 

 in the body when in its normal condition. Thus there 

 are a series of tumours, that have been included in the 



